The major goal of this revised competing renewal is the development of antigen-specific preventative therapies of type 1 diabetes that can be translated to man. Our studies to date have been aimed at a mechanistic understanding of the pathogenic processes in the NODmouse model focused on a family of class 8 protein tyrosine phosphatases that are major targets of humoral autoimmunity in humans. We have identified conserved antigenic peptide sequences that are common targets of both human and mouse CD4+ T cell clones, and conformational epitopes recognized by diabetes autoantibodies that have been conserved through over 300 million years of evolution. We have implemented new technologies for the in vitro evaluation of T cell cytokine responses (ELISPOT), and methodologies for rapidly screening pathogenic T cell receptors using retrogenic mice to further evaluate the pathogenic importance of these antigens. The grant also supported the initial characterization of humoral and cell-mediated immune responses to human Zn transporter 8, a major new ss-cell specific human diabetes autoantigen;the first to be discovered in over a decade. The current proposal while building upon these preliminary data embodies a shift in emphasis by using the NOD mouse as a pre-clinical model for the development of mechanistically-based new therapies for prevention of type 1 diabetes in humans. Specific Aim 1: Characterization of spontaneously-arising autoreactive T cells recognizing ZnT8 in NOD mice. Our approach here is largely empirical aimed at identifying multiple candidate antigen-specific TCRs and evaluating their disease potential by retrogenesis in a NOD mouse model. The fundamental objective is to identify 5-10 peptides from ZnT8 that are key to the pathogenesis of disease and which may have therapeutic potential. We will also test the diabetogenesis of the phogrin TCRs identified in the previous funding period. Specific aim 2: Pre-clinical studies of ZnT8-directed interventions in NOD mice This aim explores the practical development of ZnT8-and phogrin-based therapies, while at the same time giving heed to the mechanisms by which such suppression might occur in early and late stages of the disease. Experimentally this involves analysis of the incidence and rate of progression of T1D in NOD mice lacking ZnT8, and development of peptide and DNA based vaccination strategies based on dominant ZnT8 or phogrin T cell epitope peptides, or encoding cDNA. We anticipate that our results will confirm that both autoantigens can be effective therapeutic targets in NOD mice, both early and late during diabetogenesis, and will provide important new insight into the design of drugs and protocols that can ultimately be translated to humans.